frankenmerger/MiniLlama-1.8b-Chat-v0.1
Full fine-tune (fewer layers) vs TinyLlama/TinyLlama-1.1B-Chat-v1.0. TinyLlama/TinyLlama-1.1B-Chat-v1.0 matches part of frankenmerger/MiniLlama-1.8b-Chat-v0.1 (201 of its 363 tensors; the rest of frankenmerger/MiniLlama-1.8b-Chat-v0.1 has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 22 of 22 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 44/45, MLP 66/66, Attention 88/88
3 derivatives checked
- mradermacher/MiniLlama-1.8b-Chat-v0.1-GGUF: Same weights, converted
- mradermacher/MiniLlama-1.8b-Chat-v0.1-i1-GGUF: Quantization
- Akul/MiniLlama-1.8b-Med-safetensors: LoRA-style